Obesity is a risk factor for postmenopausal breast cancer. However, there are few studies that have evaluated the impact of prevention strategies such as calorie restriction in preclinical obesity protocols. In rodents it is easy to implement calorie restriction but in humans weight loss regimens have poor outcomes including noncompliance and weight regain leading to interest in calorie restriction mimetics including metformin. Here, we will directly compare the impact of calorie restriction to metformin treatment in a clinically relevant model of hormone responsive breast cancer. Our hypothesis is that metformin will reduce mammary tumor incidence, extend latency and reduce tumor burden to a similar degree as moderate calorie restriction. Specific Aim 1: To determine effects of metformin and calorie restriction on the development of mammary tumors, i.e., incidence, latency, tumor burden and tumor pathology in relationship to body weight. Specific Aim 2: To compare the impact of metformin and calorie restriction on longitudinal measurement of serum factors involved in mammary tumor development, i.e., insulin, glucose and IGF-I. Specific Aim 3: To assess the impact of metformin and calorie restriction on mammary epithelial cell proliferation in vivo in relationship to mammary tissue expression levels of proteins associated with proliferation and the AMPK pathway. We will use female MMTV-TGF-1/C57BL6 mice which develop mammary tumors in their second year. Mice will be fed a 33% fat diet or maintained on a low-fat normal mouse diet from 10 weeks of age (Normal Weight). Mice fed the high-fat diet will be stratified by body weight gain at 30 weeks of age into three groups as previously reported (Int.J.Obesity 28:956,2004). The heaviest mice will be designated as Obesity-Prone, the middle group as Overweight and the lightest group, Obesity-Resistant, are mice which remain in the weight range of Normal-Weight mice. This relationship is similar to what occurs in humans, i.e., diets are similar but body weights vary considerably. At 30 weeks of age mice in each body weight group will be fed Ad libitum, 25% calorie restricted or treated with metformin (100 or 200 mg/kg body weight/day) and followed until 90 weeks of age. Food intake, water consumption and body weights will be determined. Serum samples will be obtained prior to interventions and at specified time points to determine if IGF-I, insulin and/or glucose play a role in tumor development/prevention. Mammary tumor incidence, latency and tumor burden in relation to body weight and intervention will be the primary outcome. In vivo mammary cell proliferation prior to interventions and at the end of the study will be determined. Tumors and mammary tissue will be assayed for proteins associated with the AMPK pathway which is considered to be important in the actions of both calorie restriction and metformin treatment. The information obtained will provide evidence of the impact of metformin treatment and calorie restriction on mammary tumor development in relation to body weight status and will hopefully lead to human trials of obese women at high risk for breast cancer.